warwick.ac.uk/lib-publications

Original citation: Uthman, Olalekan A., Oladimeji, Olanrewaju and Nduka, Chidozie U.. (2016) Adherence to antiretroviral therapy among HIV-infected prisoners : a systematic review and meta-analysis. AIDS Care . pp. 1-9.

Permanent WRAP URL: http://wrap.warwick.ac.uk/81374 Copyright and reuse: The Warwick Research Archive Portal (WRAP) makes this work by researchers of the University of Warwick available open access under the following conditions. Copyright © and all moral rights to the version of the paper presented here belong to the individual author(s) and/or other copyright owners. To the extent reasonable and practicable the material made available in WRAP has been checked for eligibility before being made available. Copies of full items can be used for personal research or study, educational, or not-for profit purposes without prior permission or charge. Provided that the authors, title and full bibliographic details are credited, a hyperlink and/or URL is given for the original metadata page and the content is not changed in any way. Publisher’s statement: This is an Accepted Manuscript of an article published by Taylor & Francis in AIDS Care on 1 September 2016, available online: http://www.tandfonline.com/10.1080/09540121.2016.1223799 A note on versions: The version presented here may differ from the published version or, version of record, if you wish to cite this item you are advised to consult the publisher’s version. Please see the ‘permanent WRAP url’ above for details on accessing the published version and note that access may require a subscription. For more information, please contact the WRAP Team at: wrap@warwick.ac.uk

Page | 1

Adherence to antiretroviral therapy among HIV-infected prisoners: a

systematic review and meta-analysis

Olalekan A. Uthman1,2,3*, Olanrewaju Oladimeji4,5, Chidozie Nduka6

1Warwick-Centre for Applied Health Research and Delivery (WCAHRD), Warwick Medical School,

University of Warwick, Coventry, CV4 7AL, United Kingdom.

2Department of Public Health (IHCAR), Karolinska Institutet, Stockholm, Sweden.

3Centre for Evidence-Based Health Care, Stellenbosch University, Tygerberg 7505, South Africa.

4Discipline of Public Health Medicine, College of Health Science, Howard College Campus, University

of KwaZulu-Natal, Durban, South Africa.

5Center for Community Healthcare, Research and Development, Abuja, Nigeria.

6Division of Health Sciences, Warwick Medical School, University of Warwick, Coventry, UK,

*Correspondence author:

Assistant Professor Olalekan A. Uthman, Warwick-Centre for Applied Health Research and Delivery

(WCAHRD), Division of Health Sciences, Warwick Medical School, The University of Warwick,

Coventry, CV4 7AL, United Kingdom. Email: olalekan.uthman@warwick.ac.uk

Running head: ART adherence HIV-infected prisoners

Conflict of Interest/Disclosures: All authors have no conflict of interest to declare

Page | 2

ABSTRACT

Data on antiretroviral therapy (ART) adherence among prison inmates are limited and not previously

synthesised in a systematic manner. The objective of this study was to provide accurate and up-to-date

ART adherence estimates among prison inmates. We searched electronic databases for all studies

reporting adherence as a primary or secondary outcome among prison inmates. A random-effects

model was used to pool adherence rates; sensitivity, heterogeneity, and publication bias were

assessed. Eleven studies involving 2,895 HIV-infected prison inmates were included. The studies were

carried out between 1992 and 2011 and reported between 1998 and 2013. A pooled analysis of all

studies indicated a pooled estimate of 54.6% (95% confidence interval 48.1 – 60.9%) of prison inmates

had adequate (≥95%) ART adherence. The adherence estimates were significantly higher among

cross-studies and studies that used self-reported measures. In summary, our findings indicate that

optimal adherence remains a challenge among prison inmates. It is crucial to monitor ART adherence

and develop appropriate interventions to improve adherence among these population.

Keywords: adherence; prisoners; meta-analysis; antiretroviral therapy

Page | 3

INTRODUCTION

It has been documented that there is a higher burden of HIV infection among incarcerated populations

in low-, middle, and high-income countries(Jurgens, Nowak, & Day, 2011) as well as its negative impact

on continuity of care; development of trust; and, subsequently, optimal adherence(Seal, 2005).

Incarceration provides public health opportunity to provide life-saving antiretroviral therapy (ART) to

HIV-infected persons; but multiple barriers to ART access, delivery and adherence persist(R. Y. Chen

et al., 2006; Hammett, Kennedy, & Kuck, 2007; Zaller, Thurmond, & Rich, 2007). Even, after release

from prisons, non-adherence and loss-to-follow up has been reported as a major issue(Baillargeon et

al., 2009). Of paramount importance while in prison is the necessity to maintain patient confidentiality,

in order to avoid perceived and experienced stigma(Ines, Moralejo, Marcos, Fuertes, & Luna, 2008;

Pontali, 2005; Rosen et al., 2004; Small, Wood, Betteridge, Montaner, & Kerr, 2009) as well as

assisting prisoners at delivery to be linked to care in outpatient basis, maintain high level of adherence

and re-insertion in the community(Milloy et al., 2011; Springer et al., 2004; Stephenson et al., 2005).

There are limited knowledge on the level of achievable ART adherence in prison globally as well as

evidence-based interventions. We therefore conducted a systematic review with meta-analysis to fill

this research gap, to provide a more accurate and up-to-date ART adherence estimates among prison

inmates living with HIV in order to attempt to quantify the burden and inform decision regarding policy

responses and public health intervention.

METHODS

Protocol and Registration

The study background, rationale, and methods were specified in advance and documented in a

protocol to be published at the international prospective register of systematic reviews (PROSPERO;

Number: CRD42016044044)(Uthman, Nduka, & Oladimedji, 2016).

Page | 4

Eligibility Criteria

Type of studies: cross-sectional and cohort studies that reported ART adherence rates as a primary or

secondary outcome. No language, publication date or publication status restrictions were imposed. We

excluded studies that involved directly observed antiretroviral therapy. Types of participants: HIV-

infected prison inmates on ART. Types of outcome measures: adherence rates regardless of measures

(such as self-reported, pill count, etc.).

Information Sources and Search Strategy

Two of the authors (OAU & OO) conducted searches on the following electronic databases (from 1980

to January 2016): PubMed, EMBASE, SCI Web of Science, NLM Gateway and Google scholar. We

used the following keywords: “prisoners”, “jail”, "adherence", “compliance”, "antiretroviral therapy" “HIV”;

“HAART”; “ART” (see Appendix 1 for the full Medline search strategy). We searched abstract of

relevant conference proceedings from 2006 onward (the most recent ones that may not have been

indexed in NLM Gateway meeting abstracts). In addition, the bibliographies of relevant review articles

and selected articles were examined for pertinent studies.

Study selection

Two authors (OAU and OO) evaluated the eligibility of studies obtained from the literature search using

a predefined protocol, and worked independently to scan all abstracts and obtain full text of articles. In

cases of discrepancy, agreement was reached by consensus and by discussions with the third

reviewer.

Risk of bias assessment

We used the the Risk of Bias Assessment Tool for Nonrandomized Studies (RoBANS)(S. Y. Kim et al.,

2013) to appraise the risk of bias for included studies(see Appendix 2). This included information on

Page | 5

the selection bias (sample population), selection bias (participation rate), performance bias (outcome

assessment), performance bias (analytical methods to control for bias) and other form of bias. The

methodological components of the studies were assessed and classified as adequate, inadequate or

unclear. Where differences arose, they were resolved by discussions with the third reviewer.

Data abstraction

Two reviewers (OAU and OO) independently extracted and compared the data. For each study that

met the selection criteria, details were extracted on study design, study population characteristics, and

adherence measures.

Data Analysis

For the meta-analysis, we first stabilized the raw ART adherence proportions from each study using the

Freeman-Tukey variant of the arcsine square root transformed proportion(Stuart & Ord, 1994) suitable

for pooling. We used a DerSimonian-Laird random effects model(DerSimonian & Laird, 1986) due to

anticipated variations in study population, health care delivery systems and epidemic course. To

evaluate the stability of the results we applied several sensitivity analyses, including fixed effects

analysis and used a one-study removed approach.(Normand, 1999) The purpose of this analysis was

to evaluate the influence of individual studies, by estimating pooled estimate in the absence of each

study. We assessed heterogeneity among trials by inspecting the forest plots and using the chi-squared

test for heterogeneity with a 10% level of statistical significance, and using the I2 statistic where we

interpret a value of 50% as representing moderate heterogeneity.(Higgins & Thompson, 2002; Higgins,

Thompson, Deeks, & Altman, 2003) We assessed the possibility of publication bias by evaluating a

funnel plot for asymmetry. Because graphical evaluation can be subjective, we also conducted a

Egger’s regression asymmetry test(Egger, Davey, Schneider, & Minder, 1997) as formal statistical tests

for publication bias.

Page | 6

The effect of the following study-level factors on the overall adherence rates was explored using sub-

group and meta-regression analyses: type of publication (full-text versus conference abstract), study

period (earlier studies conducted before 2000 versus recent studies conducted after 2000), publication

year, study design (cross-sectional versus cohort), study’s region (North America, Europe & sub-

Saharan Africa), study size (small:<150 versus large: 150 plus), adherence threshold (≥95% versus

100%), adherence measures (self-reported versus pharmacy refill). Series of univariable random-

effects meta-regression analyses were conducted to investigate the impact of factors on the pooled

adherence proportions. Meta-analysis results were reported as combined adherence proportions with

95% confidence intervals (CIs), while meta-regression results are reported as odds ratio with 95% CIs.

We assessed the level of agreement between the review authors reviewers using kappa analysis and

reported using the Cohen kappa index (Cohen, 1960). All p-values were exact and p-value less the

0.050 was considered statistically significant. Analyses were conducted using Stata version 14 for

Windows (Stata Corp, College Station, Texas). This systematic review was reported according to the

Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines

(http://www.prisma-statement.org).(Liberati et al., 2009; Moher, Liberati, Tetzlaff, & Altman, 2009) The

PRISMA checklist is provided in the Appendix 3.

RESULTS

Search results and study characteristics

Figure 1 shows the study selection flow diagram. The literature search yielded 739 articles of which 29

duplicate records were removed. An additional 672 articles were screened by their titles and abstracts,

leaving 38 full-text articles selected for critical reading (kappa=0.72; good agreement). Twenty-seven

did not meet the inclusion criteria as no relevant outcomes were reported (Appendix 4). Eleven studies

yielding a total of 13 adherence estimates met the inclusion criteria and were included (kappa=1.00;

Page | 7

100% agreement) in the meta-analysis(Catz, Sosman, Scheuerell, & Crumble, 2002; N. E. Chen et al.,

2013; Chitsaz et al., 2013; Ines et al., 2008; Mostashari, Riley, Selwyn, & Altice, 1998; Palepu et al.,

2004; Paparizos et al., 2013; Perez et al., 2006; Soto Blanco, Perez, De Labry Lima, et al., 2005; Soto

Blanco, Perez, & March, 2005; Wakoli, Baliddawa, Kimaiyo, & Braitstein, 2010). The sample was

composed of 2895 HIV-infected prisoners on ART. Table 1 presents the characteristics of the included

studies. The studies were carried out between 1992 and 2011 and reported between 1998 and 2013.

Most were reported as journal articles (n=9, 82%); only two were presented as conference abstracts

(18%). When reported the mean age of the participants ranged from 34.0 to 44.7 years and percentage

drug users ranged from 32% to 100%. The preponderance of the studies (n=9, 82%) were cross-

sectional and two were cohort studies (18%). Most of the studies were conducted in high-income

countries (n=10, 91%) and only one study was conducted in low-income country. Studies were carried

out in the United States (n=4, 36%), Spain (n=4, 36%), Canada (n=1, 9%), Greece (n=1, 9%), and

Kenya (n=1, 9%). Most studies measured adherence using self-reported questionnaires (n=10, 91%)

only one study used pharmacy refills. Most of the studies used adherence threshold of 100% (n=8,

73%) and three studies a threshold of ≥95% (n=3, 27%).

Risk of bias of included studies

The summary risk of bias of included studies is shown in Figure 2. All studies recruited participants

from representative samples, selection bias due to sample population is low in all studies. Selection

bias due to participation rate is low in five studies, i.e. participation rate was greater than >70-85% in

these five studies and unclear in the remaining six studies. Performance bias due to outcome bias was

in all the 11 studies, they all used subjective measures (self-reported questionnaires and pharmacy

refill). Performance bias due to confounding was low in five studies that reported adjusted associations,

high in two studies and unclear in the remaining four studies. The risk of bias due to other potential bias

was low in four studies and unclear in the remaining seven studies.

Page | 8

Overall adherence to ART during and after pregnancy

Proportion of prisoners who achieved adequate adherence levels and 95% CIs from individual studies

with a pooled estimate are shown in Figure 3. The pooled ART adherence proportions for all studies

yielded an estimate of 54.6% (95% CI 48.1 to 60.9%) of patients with adequate ART adherence

(≥95%). The I2 statistics was 90.5%, indicating statistically significant heterogeneity among the studies.

The contour-enhanced funnel plot of examination of publication bias is shown in eFigure 1. The funnel

plot appears symmetric and shows no evidence of publication bias (P = 0.631 for Egger's regression

asymmetry test). The results of leave-one-study-out sensitivity analyses showed that no study had

undue influence on pooled adherence estimate (eFigure 2).

Adherence to ART by different subgroups

The results of subgroup analyses are shown in Figure 4. The pooled proportion of prisoners who

achieved adequate adherence levels was significantly higher among cross-sectional studies than

cohort studies (57.3% versus 38.1%, p-value for interaction = 0.0001); and was significantly higher

among studies that used self-reported measures than the study that used pharmacy refill (56.3%

versus 32.7%, p-value for interaction = 0.0001). However, there were no statistically significant

differential proportion of prisoners that achieved adequate adherence levels by other subgroups: type of

publication, study period, publication year, region, study size, and adherence threshold.

Factors modifying adherence estimates as identified by meta-regression analyses

Factors associated with adherence estimates and proportion of explained variability in adherence

estimates as identified by univariable meta-regression analyses are shown in Table 2. Adherence

estimates from cross-sectional studies were 94% statistically significantly higher than those from cohort

studies (OR = 1.94, 95% CI 1.43 to 12.62); and adherence estimates from studies that used self-

reported measure were 2.4 times as higher as those from pharmacy refill measure (OR = 2.44, 95% CI

Page | 9

1.60 to 3.73). Contrary to expectation, for every 10% increase in percentage of drug-users included in

the studies, the adherence estimates increased by 16% (OR = 1.16, 95% CI 1.10 to 1.29) (eFigure 3).

Percentage of drug-users, study design and adherence measure explain 40.6%, 24.9% and 20.3% in

between study variability in adherence rates respectively.

Factors associated with adherence rates as reported in individual studies

Seven studies reported factors associated with adherence estimates among prisoners (eFigure 3). The

following facilitators of optimal adherence were reported: good patients-physician and peers’

relationship, active occupation inside prison, absence of HIV symptoms, good acceptance of treatment,

and higher educational attainment. While the following barriers of optimal adherence were reported:

depressed mood, no social support, ‘bad’ quality food/food insecurity, difficulty in taking medications,

previous injecting drug use, active medical problems, alcohol use and younger age.

DISCUSSION

Main findings

To our knowledge, this is the first systematic review and meta-analysis to summarize the available data

regarding ART adherence among prison inmates and has brought together evidence from 11 studies

incorporating 2,895 prison inmates on HIV treatment. We found that the pooled proportion of prison

inmates with adequate (≥95%) ART adherence was only about 54.6%, which is still lower compared to

other high-risk subgroups, such as HIV-infected drug users (60% [95% CI 52% to 68%], 38

studies)(Malta, Magnanini, Strathdee, & Bastos, 2010), HIV-infected female sex workers (76% [95% CI

68% to 83%], 4 studies)(Mountain et al., 2014), and HIV-infected adolescents (62% [95% CI 57% to

68%], 50 studies)(S. H. Kim, Gerver, Fidler, & Ward, 2014). Interestingly, ART-adherence was twice as

high among HIV-infected prisoners in whom adherence was self-reported, compared to those for whom

adherence was determined from pharmacy re-fill records. While self-reports are a validated method for

Page | 10

measuring medication adherence(Nguyen, La Caze, & Cottrell, 2014), we cannot rule out the

disproportionately larger number of studies using this tool, as opposed to using pharamacy-refill

records, which may have biased this result in favour of the former. We also cannot rule out the potential

for recall bias when using self-reported measures.

Nonetheless, our findngs may have important public health implications. For instance, that only one out

of two HIV-infected prisoners have optimal ART adherence levels, suggests that ART non-adherence

constitutes a considerable public health burden, given potential consequences, notably antiretroviral

treatment failure and AIDS-specific mortality. Moreover, our meta-analysis was based almost entirely

on studies conducted in high-income countries. This potentially suggests that the pooled ART

adherence prevalence obtained in our study may be underestimated because prison health services in

low- and middle-income countries are less likely to be as comprehensive as prison health services in

high-income countries, such that HIV-infected prisoners in the former may be more at risk of ART non-

adherence, compared to HIV-infected prisoners in the latter.

Study limitations and strengths

The results of this meta-analysis should be interpreted with caution. We found statistically significant

heterogeneity across the studies, thus suggesting that the percentage of the variability in effect

estimates that is due to heterogeneity rather than sampling error (chance) is important. A considerable

proportion of the observed heterogeneity may be explained by differences in adherence thresholds,

proportion of participants that were drug users and study design. Nevertheless, even in the presence of

high heterogeneity, meta-analysis has been suggested as a preferred option for data synthesis

compared to qualitative or narrative interpretation; narrative synthesis can lead to misleading

conclusions that should not be generalized beyond the scope of the analysis(Ioannidis, Patsopoulos, &

Rothstein, 2008). It is worth noting that the heterogeneity observed in the current study appears to be

Page | 11

the norm rather than the exception in published ART adherence meta-analyses(Falagas, Zarkadoulia,

Pliatsika, & Panos, 2008; Peltzer & Pengpid, 2013). Another limitation is bias that can be introduced by

the methods used for measuring adherence. Most of the studies included in this meta-analysis used

self-reported adherence. Furthermore, this is a conservative bias given that self-report may

overestimate adherence, the actual levels of ART adherence may be even lower than what we are

reporting. Finally, the meta-regression analysis has several limitations. Meta-regression represents an

observational association and suffers from ecological fallacy(Thompson & Higgins, 2002). In addition,

meta-regression has low statistical power to detect an association and easily influenced by an

outlier(Lambert, Sutton, Abrams, & Jones, 2002).

Despite these limitations, the study strengths are important. We conducted comprehensive searches of

databases to ensure that all relevant, published studies were identified. We also conducted meta-

regression analyses to investigate whether any particular study-level factor explained the results and

could account for the observed variations between studies. In doing so, we have comprehensively and

robustly reviewed existing literature in this area, which points to key gaps in the current literature on

determinants of ART adherence.

CONCLUSION

Our meta-analysis showed ART adherence among prison inmates is significantly below that

recommended for adequate virologic suppression. Only about half of the prisoners achieved optimal

adherence (54.6%). Optimal adherence remains a challenge in prisoners and it is crucial to monitor

ART adherence, investigate specific risk factors for non-adherence among prisoners and develop

appropriate interventions.

Page | 12

COMPLIANCE WITH ETHICAL STANDARDS:

Funding: None

Ethical approval: This article does not contain any studies with human participants or animals

performed by any of the authors.

Informed consent: Not applicable

Page | 13

REFERENCES

Baillargeon, J., Giordano, T. P., Rich, J. D., Wu, Z. H., Wells, K., Pollock, B. H., & Paar, D. P. (2009). Accessing antiretroviral therapy following release from prison. JAMA, 301(8), 848-857. doi:10.1001/jama.2009.202

Catz, S., Sosman, J., Scheuerell, J., & Crumble, D. (2002). Antiretroviral adherence barriers in a correctional setting. Abstract no. MoPeB3259 Paper presented at the The XIV International AIDS Conference.

Chen, N. E., Meyer, J. P., Avery, A. K., Draine, J., Flanigan, T. P., Lincoln, T., . . . Altice, F. L. (2013). Adherence to HIV treatment and care among previously homeless jail detainees. AIDS & Behavior, 17(8), 2654-2666. doi:10.1007/s10461-011-0080-2

Chen, R. Y., Accortt, N. A., Westfall, A. O., Mugavero, M. J., Raper, J. L., Cloud, G. A., . . . Saag, M. S. (2006). Distribution of health care expenditures for HIV-infected patients. Clinical Infectious Diseases, 42(7), 1003-1010. doi:10.1086/500453

Chitsaz, E., Meyer, J. P., Krishnan, A., Springer, S. A., Marcus, R., Zaller, N., . . . Altice, F. L. (2013). Contribution of substance use disorders on HIV treatment outcomes and antiretroviral medication adherence among HIV-infected persons entering jail. AIDS & Behavior, 17 Suppl 2, S118-127. doi:10.1007/s10461-013-0506-0

Cohen, J. (1960). A Coefficient of Agreement for Nominal Scales. Educational and Psychological Measurement, 20(1), 37-46. doi:10.1177/001316446002000104

DerSimonian, R., & Laird, N. (1986). Meta-analysis in clinical trials. Control Clin Trials, 7(3), 177-188.

Egger, M., Davey, S. G., Schneider, M., & Minder, C. (1997). Bias in meta-analysis detected by a simple, graphical test. BMJ, 315(7109), 629-634.

Falagas, M. E., Zarkadoulia, E. A., Pliatsika, P. A., & Panos, G. (2008). Socioeconomic status (SES) as a determinant of adherence to treatment in HIV infected patients: a systematic review of the literature. Retrovirology, 5, 13. doi:10.1186/1742-4690-5-13

Hammett, T., Kennedy, S., & Kuck, S. (2007). National Survey of Infectious Diseases in Correctional Facilities: HIV and Sexually Transmitted Diseases. Report no. NCJ 217736. (Access at www.ncjrs.gov/pdffiles1/nij/grants/217736.pdf on January 6, 2016): U.S. Department of Justice.

Higgins, J. P., & Thompson, S. G. (2002). Quantifying heterogeneity in a meta-analysis. Stat.Med, 21(11), 1539-1558.

Higgins, J. P., Thompson, S. G., Deeks, J. J., & Altman, D. G. (2003). Measuring inconsistency in meta-analyses. BMJ, 327(7414), 557-560.

Ines, S. M., Moralejo, L., Marcos, M., Fuertes, A., & Luna, G. (2008). Adherence to highly active antiretroviral therapy in HIV-infected inmates. Current HIV Research, 6(2), 164-170.

Ioannidis, J. P., Patsopoulos, N. A., & Rothstein, H. R. (2008). Reasons or excuses for avoiding meta-analysis in forest plots. Bmj, 336(7658), 1413-1415. doi:10.1136/bmj.a117

Page | 14

Jurgens, R., Nowak, M., & Day, M. (2011). HIV and incarceration: prisons and detention. Journal of the International AIDS Society, 14, 26. doi:10.1186/1758-2652-14-26

Kim, S. H., Gerver, S. M., Fidler, S., & Ward, H. (2014). Adherence to antiretroviral therapy in adolescents living with HIV: systematic review and meta-analysis. AIDS, 28(13), 1945-1956. doi:10.1097/qad.0000000000000316

Kim, S. Y., Park, J. E., Lee, Y. J., Seo, H. J., Sheen, S. S., Hahn, S., . . . Son, H. J. (2013). Testing a tool for assessing the risk of bias for nonrandomized studies showed moderate reliability and promising validity. J Clin Epidemiol, 66(4), 408-414. doi:10.1016/j.jclinepi.2012.09.016

Lambert, P. C., Sutton, A. J., Abrams, K. R., & Jones, D. R. (2002). A comparison of summary patient-level covariates in meta-regression with individual patient data meta-analysis. J Clin Epidemiol, 55(1), 86-94.

Liberati, A., Altman, D. G., Tetzlaff, J., Mulrow, C., Gotzsche, P. C., Ioannidis, J. P., . . . Moher, D. (2009). The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ, 339, b2700.

Malta, M., Magnanini, M. M., Strathdee, S. A., & Bastos, F. I. (2010). Adherence to antiretroviral therapy among HIV-infected drug users: a meta-analysis. AIDS & Behavior, 14(4), 731-747. doi:10.1007/s10461-008-9489-7

Milloy, M. J., Kerr, T., Buxton, J., Rhodes, T., Guillemi, S., Hogg, R., . . . Wood, E. (2011). Dose-response effect of incarceration events on nonadherence to HIV antiretroviral therapy among injection drug users. Journal of Infectious Diseases, 203(9), 1215-1221. doi:http://dx.doi.org/10.1093/infdis/jir032

Moher, D., Liberati, A., Tetzlaff, J., & Altman, D. G. (2009). Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ, 339, b2535.

Mostashari, F., Riley, E., Selwyn, P. A., & Altice, F. L. (1998). Acceptance and adherence with antiretroviral therapy among HIV-infected women in a correctional facility. Journal of Acquired Immune Deficiency Syndromes & Human Retrovirology, 18(4), 341-348.

Mountain, E., Mishra, S., Vickerman, P., Pickles, M., Gilks, C., & Boily, M. C. (2014). Antiretroviral therapy uptake, attrition, adherence and outcomes among HIV-infected female sex workers: a systematic review and meta-analysis. PLoS ONE, 9(9), e105645. doi:10.1371/journal.pone.0105645

Nguyen, T. M., La Caze, A., & Cottrell, N. (2014). What are validated self-report adherence scales really measuring?: a systematic review. Br J Clin Pharmacol, 77(3), 427-445. doi:10.1111/bcp.12194

Normand, S. L. (1999). Meta-analysis: formulating, evaluating, combining, and reporting. Stat.Med, 18(3), 321-359.

Palepu, A., Tyndall, M. W., Chan, K., Wood, E., Montaner, J. S. G., & Hogg, R. S. (2004). Initiating highly active antiretroviral therapy and continuity of HIV care: the impact of incarceration and prison release on adherence and HIV treatment outcomes. Antiviral Therapy, 9(5), 713-719.

Paparizos, V., Kourkounti, S., Leuow, K., Georgoulas, S., Kyriakis, K., & Antoniou, C. (2013). Adherence to antiretroviral therapy among HIV-infected prisoners. Infez Med, 21(3), 189-193.

Page | 15

Peltzer, K., & Pengpid, S. (2013). Socioeconomic factors in adherence to HIV therapy in low and middle income countries Journal of Health, Population and Nutrition, 31(2), 150-170.

Perez, I. R., Soto Blanco, J. M., De Labry Lima, A. O., Castro Recio, J. M., Lopez, E. G., Basanta, J. J., & Plazaola Castano, J. (2006). Factors that affect the quality of life of prison inmates on antiretroviral treatment. AIDS Care, 18(5), 433-440. doi:10.1080/09540120500202183

Pontali, E. (2005). Antiretroviral treatment in correctional facilities. HIV Clinical Trials, 6(1), 25-37. doi:10.1310/gtqm-qrm1-fdw8-y2ft

Rosen, D. L., Golin, C. E., Schoenbach, V. J., Stephenson, B. L., Wohl, D. A., Gurkin, B., & Kaplan, A. H. (2004). Availability of and access to medical services among HIV-infected inmates incarcerated in North Carolina county jails. Journal of Health Care for the Poor & Underserved, 15(3), 413-425.

Seal, D. W. (2005). HIV-related issues and concerns for imprisoned persons throughout the world. Curr Opin Psychiatry, 18(5), 530-535. doi:10.1097/01.yco.0000179492.08064.de

Small, W., Wood, E., Betteridge, G., Montaner, J., & Kerr, T. (2009). The impact of incarceration upon adherence to HIV treatment among HIV-positive injection drug users: a qualitative study. AIDS Care, 21(6), 708-714. doi:10.1080/09540120802511869

Soto Blanco, J. M., Perez, I. R., De Labry Lima, A. O., Recio, J. M., Lopez, E. G., & Basanta, J. J. (2005). Adherence to antiretroviral treatment in prisons. AIDS Research & Human Retroviruses, 21(8), 683-688. doi:10.1089/aid.2005.21.683

Soto Blanco, J. M., Perez, I. R., & March, J. C. (2005). Adherence to antiretroviral therapy among HIV-infected prison inmates (Spain). International Journal of STD & AIDS, 16(2), 133-138.

Springer, S. A., Pesanti, E., Hodges, J., Macura, T., Doros, G., & Altice, F. L. (2004). Effectiveness of antiretroviral therapy among HIV-infected prisoners: reincarceration and the lack of sustained benefit after release to the community. Clinical Infectious Diseases, 38(12), 1754-1760. doi:10.1086/421392

Stephenson, B. L., Wohl, D. A., Golin, C. E., Tien, H. C., Stewart, P., & Kaplan, A. H. (2005). Effect of release from prison and re-incarceration on the viral loads of HIV-infected individuals. Public Health Reports, 120(1), 84-88.

Stuart, A., & Ord, J. K. (1994). Kendall's Advanced Theory of Statistics (Vol. 6th ed). London, England: Arnold Publishers.

Thompson, S. G., & Higgins, J. P. (2002). How should meta-regression analyses be undertaken and interpreted? Stat Med, 21(11), 1559-1573. doi:10.1002/sim.1187

Uthman, O. A., Nduka, C. N., & Oladimedji, O. (2016). Olalekan Uthman, Chidozie Nduka, Olanrewaju Oladimeji. Adherence to antiretroviral therapy among HIV-infected prisoners: a systematic review and meta-analysis. PROSPERO 2016:CRD42016044044 Available from http://www.crd.york.ac.uk/PROSPERO/display_record.asp?ID=CRD42016044044.

Wakoli, A., Baliddawa, J., Kimaiyo, S., & Braitstein, P. (2010). Adherence challenges to combination antiretroviral therapy (cART) among HIV- positive prisoners at a high security prison, Western Kenya. abstract no. TUPE0173. Paper presented at the XVIII International AIDS Conference Vienna, Austria.

Page | 16

Zaller, N., Thurmond, P., & Rich, J. D. (2007). Limited spending: an analysis of correctional expenditures on antiretrovirals for HIV-infected prisoners. Public Health Reports, 122(1), 49-54.

Page | 17

TABLES

Table 1: Characteristics of included studies Author (year) Type of

publication

Study

period

Study

design

cou

ntry

Income

category

Male

(%)

Mean

age

Drug

users

(%)

Sampl

e size

Adher

ence

measu

re

Adher

ence

thres

hold

Catz et al. 2002(Catz et al., 2002) Conference

e abstract

Not

reporte

d

Cross-

sectional

USA High-

income

92 Not

reporte

d

Not

reported

50 Self-

reporte

d

100

Chen et al. 2013(N. E. Chen et al., 2013) Full-text 2007-

2010

Cross-

sectional

USA High-

income

66.2 44.7 78.5 653 Self-

reporte

d

95

Chitsaz et al. 2013(Chitsaz et al., 2013) Full-text 2007-

2011

Cross-

sectional

USA High-

income

72.3 42.8 72.3 1163 Self-

reporte

d

95

Ines et al. 2008(Ines et al., 2008) Full-text 1993-

1995

Cross-

sectional

Spai

n

High-

income

92 NR 72 50 Self-

reporte

d

100

Mostashari et al. 1998(Mostashari et al.,

1998)

Full-text 1993-

1995

Cross-

sectional

USA High-

income

0 35 90 102 Self-

reporte

d

100

Palepu et al. 2004(Palepu et al., 2004) Full-text 1997-

2002

Cohort Can

ada

High-

income

90 34 44.6 101 Pharm

Refills

100

Paparizos et al. 2013(Paparizos et al.,

2013)

Full-text 2001-

2011

Cohort Gre

ece

High-

income

90.3 37.45 32.2 93 Self-

reporte

d

95

Perez et al 2006(Perez et al., 2006) Full-text 2003 Cross-

sectional

Spai

n

High-

income

88.7 35.7 Not

reported

160 Self-

reporte

d

100

Soto Blanco et al 2005 (a)(Soto Blanco,

Perez, & March, 2005)

Full-text 2000 Cross-

sectional

Spai

n

High-

income

84.7 Not

reporte

d

100 177 Self-

reporte

d

100

Soto Blanco et al 2005 (b)(Soto Blanco,

Perez, De Labry Lima, et al., 2005)

Full-text 2002 Cross-

sectional

Spai

n

High-

income

90 35.5 94 281 Self-

reporte

d

100

Wakoli et al. 2010(Wakoli et al., 2010) Conference

e abstract

Not

reporte

d

Cross-

sectional

Ken

ya

Low-

income

80 35 Not

reported

65 Self-

reporte

d

100

Page | 18

Table 2: Factors associated with adherence estimates identified by meta-regression analyses Factors OR (95% CI) p-value R2 (%)

Full-text (vs. abstract) 0.63 (0.25 to 1.62) 0.305 1.0

Publication year 0.99 (0.92 to 1.06) 0.779 0.0

Recent (vs. earlier) studies 1.10 (0.90 to 1.35) 0.310 1.7

Cross-sectional (vs cohort) study 1.94 (1.43 to 2.62) 0.000 24.9

American (vs European) study 1.05 (0.51 to 2.17) 0.888 0.0

Male (%) 0.99 (0.98 to 1.01) 0.386 0.0

Mean age (years) 1.03 (0.92 to 1.14) 0.553 0.0

Drug-user (per 10% increase) 1.16 (1.10 to 1.29) 0.000 40.6

Large (vs small) study 1.25 (0.65 to 2.45) 0.461 0.0

100% (vs. 95%) threshold 1.07 (0.54 to 2.12) 0.836 0.0

Self-reported (vs. pharm refill) 2.44 (1.60 to 3.73) 0.000 20.3

FIGURE LEGENDS

Figure 1: Study selection flow diagram

Figure 2: Risk of bias included studies

Figure 3: Pooled proportion of prison inmates’ adherent to antiretroviral therapy

Figure 4: Pooled proportion of prison inmates’ adherent to antiretroviral therapy, by different sub-groups

SUPPLEMENTARY DIGITAL CONTENT

Contents

APPENDICIES .............................................................................................................................................. 24

Appendix 1: Medline search strategy ............................................................................................................ 24

Appendix 2: Risk of bias assessment ........................................................................................................... 26

Appendix 3: PRISMA checklist .................................................................................................................... 27

Appendix 4: Excluded studies ...................................................................................................................... 29

ONLINE ONLY FIGURES ............................................................................................................................... 33

eFigure 1: Contour-enhanced funnel plot ...................................................................................................... 33

eFigure 2: Leave-one-out sensitivity analyses ............................................................................................... 34

eFigure 3: Association between percentage drug users and adherence rate among prison inmates ................... 35

eFigure 4: Overview of factors associated with adherence as reported by individual studies .............................. 36

24

APPENDICIES

Appendix 1: Medline search strategy 1 adherence.mp. 2 nonadherence.mp. 3 compliance.mp. 4 noncompliance.mp. 5 non-compliance.mp. 6 "pill count?".mp. 7 MEMS.mp. 8 "medication event monitoring system".mp. 9 "pharm* refil*".mp. 10 exp Medication Adherence/ 11 or/1-10 12 "*prison*".mp. 13 prisoner?.mp. 14 imprison.mp. 15 incarcerat*.mp. 16 "*offend*".mp. 17 "remand*".mp. 18 "*detain*".mp. 19 "*criminal*".mp. 20 "*convict*".mp. 21 "*felon*".mp. 22 pre-trial.mp. 23 under-trial.mp. 24 "jail".mp. 25 "gaol".mp. 26 "detention".mp. 27 "correction*".mp. 28 "sentence*".mp. 29 "probation*".mp. 30 "parole*".mp. 31 re-entry.mp. 32 reentry.mp. 33 "post-release".mp. 34 "transition*".mp. 35 "supervis*".mp. 36 inmate?.mp. 37 in-mate?.mp. 38 "correctional facility".mp. 39 exp prisoner/ 40 exp prison/ 41 exp criminal/ 42 exp detention/ 43 exp jail/ 44 exp parole/ 45 exp offender/ 46 or/12-45

25

47 (hiv or hiv or hiv-1 or hiv-2).mp. 48 human immunodeficiency virus.mp. 49 human immunedeficiency virus.mp. 50 human immune-deficiency virus.mp. 51 hiv infections.mp. 52 aids.ti,ab. 53 acquired immune-deficiency syndrome.mp. 54 acquired immunedeficiency syndrome.mp. 55 acquired immunodeficiency syndrome.mp. 56 acquired immuno-deficiency syndrome.mp. 57 (HAART or highly active anti?retroviral therapy or highly active anti retroviral therapy).mp. 58 Antiretroviral Therapy.mp. 59 retroviral*.mp. 60 Antiviral Agents.mp. 61 human immunodeficiency.mp. 62 antiretroviral*.mp. 63 exp hiv infections/ 64 exp human immunodeficiency virus/ 65 exp acquired immune-deficiency syndrome/ 66 exp acquired immunodeficiency syndrome/ 67 exp acquired immuno-deficiency syndrome/ 68 exp hiv/ 69 exp hiv-1/ 70 exp hiv-2/ 71 exp HIV SERONEGATIVITY/ 72 exp HIV SEROPOSITIVITY/ 73 exp HIV SEROPREVALENCE/ 74 exp HIV ANTIBODIES/ 75 exp ANTIRETROVIRAL THERAPY, HIGHLY ACTIVE/ 76 exp Antiretroviral Therapy/ 77 or/47-76 78 11 and 46 and 77

26

Appendix 2: Risk of bias assessment

Bias type Adequate Inadequate Unclear

Selection (sample

population)

Rationale for cases

(prisoners) selection

explained

Sample selection

ambiguous and sample

unlikely to be

representative

Insufficient

information

Selection

(participation rate)

High participation rate (>70-

85%)

Low participation rate

(<70%)

Insufficient

information

Performance bias

(outcome

assessment)

Objective measures of

adherence

Self-reported measure of

adherence

Insufficient

information

Performance bias

(analytical methods

to control for bias)

Analysis appropriate for type

of sample (unadjusted,

univariable analyses etc.)

Analysis does not

account for common

adjustment (adjusted,

multivariable analyses)

Insufficient

information

Other form of bias There is no evidence of bias

from other sources.

There is potential bias

present from other

sources

Insufficient

information

27

Appendix 3: PRISMA checklist

Section/topic # Checklist item Reported on page #

TITLE

Title 1 Identify the report as a systematic review, meta-analysis, or both. 1

ABSTRACT

Structured summary

2 Provide a structured summary including, as applicable: background; objectives; data sources; study eligibility criteria, participants, and interventions; study appraisal and synthesis methods; results; limitations; conclusions and implications of key findings; systematic review registration number.

2

INTRODUCTION

Rationale 3 Describe the rationale for the review in the context of what is already known.

3

Objectives 4 Provide an explicit statement of questions being addressed with reference to participants, interventions, comparisons, outcomes, and study design (PICOS).

3

METHODS

Protocol and registration

5 Indicate if a review protocol exists, if and where it can be accessed (e.g., Web address), and, if available, provide registration information including registration number.

NA

Eligibility criteria 6 Specify study characteristics (e.g., PICOS, length of follow-up) and report characteristics (e.g., years considered, language, publication status) used as criteria for eligibility, giving rationale.

3

Information sources

7 Describe all information sources (e.g., databases with dates of coverage, contact with study authors to identify additional studies) in the search and date last searched.

4

Search 8 Present full electronic search strategy for at least one database, including any limits used, such that it could be repeated.

4

Study selection 9 State the process for selecting studies (i.e., screening, eligibility, included in systematic review, and, if applicable, included in the meta-analysis).

4

Data collection process

10 Describe method of data extraction from reports (e.g., piloted forms, independently, in duplicate) and any processes for obtaining and confirming data from investigators.

5

Data items 11 List and define all variables for which data were sought (e.g., PICOS, funding sources) and any assumptions and simplifications made.

5

Risk of bias in individual studies

12 Describe methods used for assessing risk of bias of individual studies (including specification of whether this was done at the study or outcome level), and how this information is to be used in any data synthesis.

4

Summary measures

13 State the principal summary measures (e.g., risk ratio, difference in means).

5-6

28

Section/topic # Checklist item Reported on page #

Synthesis of results

14 Describe the methods of handling data and combining results of studies, if done, including measures of consistency (e.g., I2) for each meta-analysis.

5-6

Risk of bias across studies

15 Specify any assessment of risk of bias that may affect the cumulative evidence (e.g., publication bias, selective reporting within studies).

4

Additional analyses

16 Describe methods of additional analyses (e.g., sensitivity or subgroup analyses, meta-regression), if done, indicating which were pre-specified.

5-6

RESULTS

Study selection 17 Give numbers of studies screened, assessed for eligibility, and included in the review, with reasons for exclusions at each stage, ideally with a flow diagram.

6

Study characteristics

18 For each study, present characteristics for which data were extracted (e.g., study size, PICOS, follow-up period) and provide the citations.

6-7

Risk of bias within studies

19 Present data on risk of bias of each study and, if available, any outcome level assessment (see item 12).

7

Results of individual studies

20 For all outcomes considered (benefits or harms), present, for each study: (a) simple summary data for each intervention group (b) effect estimates and confidence intervals, ideally with a forest plot.

7

Synthesis of results

21 Present results of each meta-analysis done, including confidence intervals and measures of consistency.

7-8

Risk of bias across studies

22 Present results of any assessment of risk of bias across studies (see Item 15).

7

Additional analysis 23 Give results of additional analyses, if done (e.g., sensitivity or subgroup analyses, meta-regression [see Item 16]).

8-9

DISCUSSION

Summary of evidence

24 Summarize the main findings including the strength of evidence for each main outcome; consider their relevance to key groups (e.g., healthcare providers, users, and policy makers).

9

Limitations 25 Discuss limitations at study and outcome level (e.g., risk of bias), and at review-level (e.g., incomplete retrieval of identified research, reporting bias).

10-11

Conclusions 26 Provide a general interpretation of the results in the context of other evidence, and implications for future research.

11

FUNDING

Funding 27 Describe sources of funding for the systematic review and other support (e.g., supply of data); role of funders for the systematic review.

11

29

Appendix 4: Excluded studies

Study Reason

Anonymous1 No relevant outcome reported

Babudieri 20002 No relevant outcome reported

Babudieri 20053 No relevant outcome reported

Baillargeon 20004 No relevant outcome reported

Baillargeon 20095 No relevant outcome reported

Beckwith 20106 No relevant outcome reported

Bird 19937 No relevant outcome reported

De Groot 20068 Narrative review

Fontana 20079 No relevant outcome reported

Frank 199910 Narrative review

Gallego 200311 No relevant outcome reported

Gir 200512 No relevant outcome reported

Kantrowitz 200513 No relevant outcome reported

Karus 200714 No relevant outcome reported

Pai 200915 No relevant outcome reported

Pontali 200516 Narrative review

Roberson 200917 No relevant outcome reported

Saberi 201218 No relevant outcome reported

Scheyett 200819 No relevant outcome reported

Seal 200520 Narrative review

Small 200921 No relevant outcome reported

Spaulding 200922 No relevant outcome reported

Springer 2004 23 No relevant outcome reported

Springer 2005 24 Narrative review

Springer 201125 Narrative review

Westergaard 2011 26 No relevant outcome reported

White 2006 27 No relevant outcome reported

1. Mean streets: out of prison, out of HIV med compliance. Prisoner re-entry needs targeted

focus. (2009). AIDS alert, 24(7), 73-75.

2. Babudieri, S., Aceti, A., D'Offizi, G. P., Carbonara, S., & Starnini, G. (2000). Directly observed therapy to treat HIV infection in prisoners. JAMA, 284(2), 179-180.

3. Babudieri, S., Pintus, A., Maida, I., Starnini, G., & Rezza, G. (2005). Does counseling increase sustained benefit of HAART among prison inmates after release to the community? Clinical Infectious Diseases, 40(2), 321-322; author reply 322-323.

30

4. Baillargeon, J., Borucki, M. J., Zepeda, S., Jenson, H. B., & Leach, C. T. (2000). Antiretroviral prescribing patterns in the Texas prison system. Clinical Infectious Diseases, 31(6), 1476-1481. doi:10.1086/317478

5. Baillargeon, J., Giordano, T. P., Rich, J. D., Wu, Z. H., Wells, K., Pollock, B. H., & Paar, D. P. (2009). Accessing antiretroviral therapy following release from prison. JAMA, 301(8), 848-857.

6. Beckwith, C. G., Zaller, N. D., Fu, J. J., Montague, B. T., & Rich, J. D. (2010). Opportunities to diagnose, treat, and prevent HIV in the criminal justice system. Journal of Acquired Immune Deficiency Syndromes: JAIDS, 55 Suppl 1, S49-55.

7. Bird, A. G., Gore, S. M., Burns, S. M., & Duggie, J. G. (1993). Study of infection with HIV and related risk factors in young offenders' institution. BMJ, 307(6898), 228-231.

8. De Groot, A. S., Dilorenzo, M., Sylla, M., & Bick, J. (2006). Challenges and opportunities for HIV care in jails and prisons in the United States. International Journal of Prisoner Health, 2(3), 173-191.

9. Fontana, L., & Beckerman, A. (2007). Recently released with HIV/AIDS: primary care treatment needs and experiences. Journal of Health Care for the Poor & Underserved, 18(3), 699-714.

10. Frank, L. (1999). Prisons and public health: emerging issues in HIV treatment adherence. Journal of the Association of Nurses in AIDS Care, 10(6), 24-32.

11. Gallego, O., Corral, A., De Mendoza, C., Gonzalez-Lahoz, J., & Soriano, V. (2003). High rate of resistance to antiretroviral drugs among HIV-infected prison inmates. Med Sci Monit, 9(6), Cr217-221.

12. Gir, E., Vaichulonis, C. G., & de Oliveira, M. D. (2005). [Adhesion to anti-retroviral therapy by individuals with HIV/AIDS seen at an institution in the interior of Sao Paulo]. Revista Latino-Americana de Enfermagem, 13(5), 634-641.

13. Kantrowitz Kunkel, J., DeVissi, B., Golin, C., Kaplan, A., McKay, T. E., Wohl, D., & Stephenson, B. (2005). Social determinants of medication adherence and general health among HIV-positive prison inmates. American Public Health Association 133rd Annual Meeting & Exposition; Dec 10 2005; Philadelphia,MA. Retrieved from http://onlinelibrary.wiley.com/o/cochrane/clcentral/articles/654/CN-00635654/frame.html

14. Karus, D., Raveis, V. H., Ratcliffe, M., Rosefield, H. A., Jr., & Higginson, I. J. (2007). Health status and service needs of male inmates seriously ill with HIV/AIDS at two large urban jails. American Journal of Mens Health, 1(3), 213-223. doi:http://dx.doi.org/10.1177/1557988307304230

15. Pai, N. P., Estes, M., Moodie, E. E. M., Reingold, A. L., & Tulsky, J. P. (2009). The impact of antiretroviral therapy in a cohort of HIV infected patients going in and out of the San Francisco county jail. PLoS ONE [Electronic Resource], 4(9), e7115.

31

16. Pontali, E. (2005). Antiretroviral treatment in correctional facilities. HIV Clinical Trials, 6(1), 25-37.

17. Roberson, D. W., White, B. L., & Fogel, C. I. (2009). Factors influencing adherence to antiretroviral therapy for HIV-infected female inmates. Journal of the Association of Nurses in AIDS Care, 20(1), 50-61.

18. Saberi, P., Caswell, N. H., Jamison, R., Estes, M., & Tulsky, J. P. (2012). Directly observed versus self-administered antiretroviral therapies: preference of HIV-positive jailed inmates in San Francisco. Journal of Urban Health, 89(5), 794-801.

19. Scheyett, A., Parker, S., Golin, C., White, B., Davis, C. P., & Wohl, D. (2010). HIV-infected prison inmates: depression and implications for release back to communities. AIDS & Behavior, 14(2), 300-307.

20. Seal, D. W. (2005). HIV-related issues and concerns for imprisoned persons throughout the world. Curr Opin Psychiatry, 18(5), 530-535.

21. Small, W., Wood, E., Betteridge, G., Montaner, J., & Kerr, T. (2009). The impact of incarceration upon adherence to HIV treatment among HIV-positive injection drug users: a qualitative study. AIDS Care, 21(6), 708-714.

22. Spaulding, A. C., Seals, R. M., Page, M. J., Brzozowski, A. K., Rhodes, W., & Hammett, T. M. (2009). HIV/AIDS among inmates of and releasees from US correctional facilities, 2006: declining share of epidemic but persistent public health opportunity. PLoS ONE [Electronic Resource], 4(11), e7558.

23. Springer, S. A., & Altice, F. L. (2005). Managing HIV/AIDS in correctional settings. Current HIV/AIDS Reports, 2(4), 165-170.

24. Springer, S. A., Azar, M. M., & Altice, F. L. (2011). HIV, alcohol dependence, and the criminal justice system: a review and call for evidence-based treatment for released prisoners. American Journal of Drug & Alcohol Abuse, 37(1), 12-21.

25. Springer, S. A., Pesanti, E., Hodges, J., Macura, T., Doros, G., & Altice, F. L. (2004). Effectiveness of antiretroviral therapy among HIV-infected prisoners: reincarceration and the lack of sustained benefit after release to the community. Clinical Infectious Diseases, 38(12), 1754-1760.

26. Westergaard, R. P., Kirk, G. D., Richesson, D. R., Galai, N., & Mehta, S. H. (2011). Incarceration predicts virologic failure for HIV-infected injection drug users receiving antiretroviral therapy. Clinical Infectious Diseases, 53(7), 725-731.

27. White, B. L., Wohl, D. A., Hays, R. D., Golin, C. E., Liu, H., Kiziah, C. N., . . . Kaplan, A. H. (2006). A pilot study of health beliefs and attitudes concerning measures of antiretroviral adherence among prisoners receiving directly observed antiretroviral therapy. AIDS Patient Care & Stds, 20(6), 408-417.

32

33

ONLINE ONLY FIGURES

eFigure 1: Contour-enhanced funnel plot

34

eFigure 2: Leave-one-out sensitivity analyses

35

eFigure 3: Association between percentage drug users and adherence rate among prison inmates

36

eFigure 4: Overview of factors associated with adherence as reported by individual studies

37